You are on page 1of 10


Education and
It’s not just going to school, but learning

ven before and certainly ever since the Stuck in the Middle (Figure 1) U.S. students again fared poorly on the
1983 release of A Nation at Risk by the
National Commission on Excellence in
Education, national economic competitive-
ness has been offered as a primary reason
for pushing school reform. The commission
warned,“If only to keep and improve on the
slim competitive edge we still retain in world
markets, we must dedicate ourselves to the
reform of our educational system for the
benefit of all—old and young alike, affluent
and poor, majority and minority.” Respond-
ing to these urgent words, the National Gov-
ernors Association, in 1989, pledged that U.S.
students would lead the world in math and CAN
science achievement by 2000.
According to the latest international math
and science assessment conducted by the
Organisation of Economic Co-operation and
Development’s (OECD) Programme for
International Student Assessment (PISA) (see
Figure 1), the United States remains a long USA
distance from that target. Rather than wor-
rying about the consequences, some have
begun to question what all the fuss was about.
Education researcher Gerald Bracey, for MEX
example, has argued that no one has “pro-
vided any data on the relationship between
the economy’s health and the performance of most systematically about the way in which
schools. Our long economic boom suggests “human capital” affects a nation’s economic
there isn’t one—or that our schools are bet- future, have skirted the heart of the question
ter than the critics claim.” by looking only at “school attainment,”
Truth be told, the Bracey critique is not namely the average number of years students
entirely misplaced. Most commentators rely remain in school.
more on the commonsense understanding Using average years of schooling as an indi-
that countries must have good schools to cator of a country’s human capital has at least
succeed economically rather than presenting two major drawbacks. First and foremost, the CHI
conclusive empirical evidence that connects approach assumes that students in diverse URU
what students learn in school to what sub- school systems around the world receive the
sequently happens in a nation’s economy. same educational benefits from a year of
Even economists, the people who think the schooling. A year of schooling in Papua New

62 E D U C AT I O N N E X T / S P R I N G 2 0 0 8

org S P R I N G 2 0 0 8 / E D U C AT I O N N E X T 63 . PISA. Economic Growth something while there that matters latest international test of math. placing 21st among the 32 OECD countries. ISL SWE FIN RUS NOR EST DEN LAT NED LTU BEL IRL POL GBR GER LUX SUI LIE CZE SVK AUT HUN FRA SLO CRO ROU SRB BUL POR ESP ITA AZE TUR KOR GRE JPN ISR TPE HKG THA MAC Mean Math Score. 2006 510-549 490-510 415-489 AUS Below 415 No Data NZL SOURCE: Programme for International Student Assessment (PISA) 2006 www.educationnext.

we could assess test. ical. enhancing human capital will become increasingly crit- countries over a long enough period of time to allow for sys. agencies have made no attempt to link their results to one ized tests as rough measures of the average level of cognitive another. Beginning in the mid. most sophisticated. and because testing grade levels. At one becomes increasingly interdependent or “flat. Second. Now that test- score data for many countries over an extended period of time are readily available.educationnext. for example. informative measures of school attainment.” to use New time. Increasing the average number of years of school. economy. This required a norm against which each test could be how human capital relates to differences in economic growth calibrated. something has to be learned there. The cognitive skills. tries at various points in time. we could construct that norm by for 50 countries from 1960 to 2000. With that information in hand. first step was to use the 12 PISA and other international math In a series of studies conducted over several years. we needed to develop comparable scores for each skill in a given country. As the world continues to change. some. this measure does not account for learn. ing that takes place outside the classroom—within families. The level of cognitive skills of (NAEP). We used performance on 12 of these standard. As the world skills” among those entering a country’s work force. to construct four of us have explored the role of both school attainment an index of cognitive skill levels for a large sample of coun- and cognitive skills in economic growth. the and science assessments. The more open the performance of students on tests in math and science. something that might be called the average level of “cognitive skills” among those entering a country’s work force. dating back to 1964. With this information. we used the variation in increased levels of school attainment also boost cognitive scores across a subset of the more-advanced developed 64 E D U C AT I O N N E X T / S P R I N G 2 0 0 8 www. it is not enough simply to spend more productive. more countries over a using information from tests in the United States. Because the number of coun- 1960s. tries participating in the 12 test administrations changed dents’ performance in mathematics and science at various from one administration to the next. the National Assessment of Educational Progress expressed in A Nation at Risk. internationally comparable information on student York Times columnist Thomas Friedman’s familiar terminol- performance was not available for a sufficient number of ogy. it has also maintained a separate longitudinal testing system What we discovered gives credence to the concerns of its own. which is why economists relied upon the less ill afford to rest easily on its past accomplishments. to one another via the connection of those tests to the ing attained by the labor force boosts the economy only when NAEP. We also discovered that the size of the impact of cognitive among peers. the coun- longer period of time than any previous study. or via the Internet. skills depends on whether a nation’s economy is open to out- A more direct measure of a country’s human capital is the side trade and other external influences. Fortunately.Guinea and a year of schooling in Japan are treated as equally skills. We were also try that has had the earliest. it is possible to supplement measures of Measuring Cognitive Skills educational attainment with these more direct measures of Reaching these conclusions required a multistep analysis. it was possible to a nation’s students has a large effect on its subsequent economic calibrate scores on each of the separate international tests growth rate. and most A more direct measure of a country’s human capital is the performance of students on tests in math and science. To obtain further precision. and protection of property rights. international agencies started conducting tests of . able to pay close attention to institutional factors that influ. the United States can tematic study. comprehensive system of testing. time in school. such as openness of the economy and participated in all of the international tests since 1964. The United States has ence economic growth. the more important it is that a country’s students thing that might be called the average level of “cognitive are acquiring high levels of cognitive skills. In other words.

Estonia Germany Czech Rep. all international tests administered between 1964 and 2003 have been placed on a common S P R I N G 2 0 0 8 / E D U C AT I O N N E X T 65 . Jordan United Kingdom Italy Jordan Malaysia Turkey Singapore Iceland Nigeria Latvia Uruguay 400s Canada Yugoslavia Venezuela Poland Thailand New Zealand Malaysia Kuwait Spain Macedonia Italy Denmark Tunisia Italy Colombia Israel Latvia Norway Iran United States Spain Russian Fed. Rep. Notice that students in the United States. Japan Korea. Countries with higher scores within a category appear in the left-hand column. math. research GROWTH HANUSHEK ET AL. 1960s-1970s 1980s 1990s 2000s Israel Japan Taiwan Finland New Zealand Netherlands Korea. allowing readers to identify changes in test performances over time. Bahrain Thailand Greece Slovenia Argentina Luxembourg Lithuania Greece Palestine Swaziland Thailand Bulgaria Mexico Nigeria Portugal Romania Egypt Philippines Israel Portugal Malawi Turkey Chile Philippines Indonesia Lebanon Saudi Arabia Zimbabwe Kuwait Morocco 300s Chile Indonesia Botswana Botswana Brazil Belize Philippines Tunisia Peru Albania South Africa Ghana 200s Morocco South Africa Note: Countries are ordered by average test score received in one or more tests in reading. SOURCE: Authors www. Rep. France Italy Finland Switzerland Romania Lithuania Moldova United States France United States Cyprus Hungary Luxembourg Thailand Norway France Trinidad&Tobago United States Israel Chile Sweden New Zealand Moldova Iceland Cyprus India Australia Germany Macedonia Germany Armenia Iran Poland Norway Iran Denmark Serbia Belgium Ireland Colombia Slovak Rep. while students in the Netherlands and Finland have improved. Germany and Hungary have slipped. Rep. Korea.educationnext. Liechtenstein Netherlands Switzerland Hungary Sweden Sweden New Zealand United Kingdom United Kingdom Belgium Belgium Canada Austria Czech Rep. Comparing Cognitive Skills (Figure 2) For the first time. Singapore Australia Australia Macao-China Finland Slovenia Canada United Kingdom Hong Kong Australia 500s Sweden Bulgaria Netherlands Netherlands Austria Ireland Russian Fed. and science within the indicated decades. Japan Taiwan France Hungary Singapore Japan Belgium Hong Kong Finland Hong Kong Hungary Slovak Rep.

one full percentage point higher annually over the following 40-year period than the second country’s Impact on Economic Growth growth rate.educationnext. level of economic development.. ity to economic growth lies in our ability to explain global vari- First.5 standard devi- lower levels of economic development. Further.Lifting it by 0. such top-performing countries as Finland and Hong Kong formers that include Albania. the initial level of economic development. by the performance of students on math and science tests. In Figure 2. again. to aggregate all available scores for each country into mea. is considerably larger. little less than the current difference in the scores between formers such as the United States and Germany. Specifically. age schooling in a country increased the average 40-year Of course. i. When we performed the analysis parable measure of cognitive skill levels include the 30 democ. we looked just at the impact of average school attain. and cognitive skills are not the only That may not seem like much. as other statistical models of economic growth. (An adjustment was made nomic growth with information only about school attainment for the initial level of GDP because it is “easier” to grow if you levels and the level of a country’s GDP in 1960. the Philippines. average per. each additional year of aver. once the impact of higher levels of We wanted to use this new information to compare the eco. ation in GDP growth. cognitive skills are taken into account. But when we also included cognitive skills in our own. this time also including the average test-score perfor- racies that have market economies and have been accepted as mance of a country in our model. The other 20 countries are at If one country’s test-score performance was 0. higher levels of cognitive higher annual rate over subsequent decades. ferences in economic growth. In other words. we found that countries members of the OECD. schooling attainment. By following these two steps. across the 50 countries. But the impact of improved cognitive skills. that is. we skill appear to play a major role in explaining international dif- found that. it is easier to copy more to explain only one-quarter of the differences we saw among productive technologies than to initiate progress on your countries. as measured sures of average cognitive skill levels for each country.) When we performed this analysis. we were able would otherwise have occurred. In other words.e. and low per. When we tried to account for eco- ment on the economic growth rate. A country ben- took measures of average educational attainment and average efits from asking its students to remain in school for a longer cognitive skill levels for as many countries as possible and period of time only if the students are learning something examined their relationship to the average annual growth as a consequence.countries to obtain an estimate of the spread in scores is a boost to annual growth rates of more than 10 percent of what across countries. The 50 countries for which we were able to develop a com. and South Africa. dwindles to nothing (see Figure 3). Could it be that some World War II. additional years of the benefits of higher levels of cognitive skills. a significant amount. on average. but consider the fact that since factors that affect economic growth. the world economic growth rate has been around other factor we have overlooked is responsible for the close 2 to 3 percent of GDP . the economy grew at a these three factors.37 percentage points connection between test scores and economic growth? 66 E D U C AT I O N N E X T / S P R I N G 2 0 0 8 www. most of which are at a relatively high with higher test scores experienced far higher growth rates. we were able are starting out at a lower level. we were able to attribute economists before us. higher levels of cognitive skill appear to play of schooling in a country was higher. and the United States—the first country’s growth rate was.37 percentage points. We therefore schooling. the significance for eco- nomic benefits of higher levels of just school attainment with nomic growth of school attainment. growth rate in GDP by about 0. we found. that when the average number of years nearly three-quarters of the differences among countries to A country benefits from asking its students to remain in school for a consequence. you can ations higher than another country during the 1960s—a identify top performers like Finland and Japan. rate in the country’s gross domestic product (GDP) per capita Another indication of the importance of education qual- from 1960 through 2000.

0 2. proxies. neither of these additional of its property rights and its openness to international trade. which is represented by a dot. Explaining Economic Growth (Figure 3) How much students learn.0 -3.5 -4. of cognitive skills on annual economic growth becomes some.0 -1. when we took into One of our tests was particularly interesting. longer period of time only if the students are learning something as a a major role in explaining international differences in economic growth. adjusted for The y-axis indicates growth rates from 1960 and 2000.0 1.0 -1.educationnext.0 1.0 3. We performed a variety of additional tests to assess the what smaller. This is the best available estimate relationships over shorter periods of time. however.5 -2.5 United States 0.0 -3. The x-axis shows school attainment adjusted for test for school attainment. The United States. The solid line plots the relationship between the two scores. When those two factors are taken into S P R I N G 2 0 0 8 / E D U C AT I O N N E X T 67 .0 -4.5 -1. the positive effect Once again.0 -2. The solid line plots the relationship between the two variables variables among the 50 countries with available test score information. variables was significantly associated with economic growth.0 1. adjusted for GDP in GDP in 1960 and school attainment.0 -2. the strong effect of cognitive skills remained clear. we estimated the dard deviation of test scores.0 0.63 percentage points per half of a stan.0 -1.0 0. each of each of which is represented by a dot. SOURCE: Authors Other economic research has identified two additional fac. The x-axis shows test scores adjusted 1960 and test scores. is the key to economic growth. has had a higher growth rate during this period than would be expected given its test scores and levels of school attainment. but is still 0.0 1.0 0. used different of the size of the impact of cognitive skills on economic growth.0 -1. among the 50 countries with available test score information.0 0.5 Adjusted growth rate Adjusted growth rate United States 2.0 -0. such as latitude or the fraction of the land area of a tors that affect a country’s economic growth rate: the security country that is located in the tropics. not how long they stay in school. Test scores and economic growth Years of education and economic growth 4.5 1. For example. We thought account the total fertility rate and common geographical it possible that the effect of cognitive skills could be the result www. validity of these basic results.0 3. research GROWTH HANUSHEK ET AL. However.0 4. growth are fertility and geography.0 0.0 5.0 2.0 -1. subsets of international tests. and compared smaller groups Other commonly discussed determinants of economic of the 50 countries.0 Adjusted test score Adjusted years of schooling Note: The y-axis indicates growth rates from 1960 and 2000.

To show a country’s ability to develop a large cadre growing economies. The reasons that a substantial cadre of highly skilled citi- zens and near-universal basic skills matter are not difficult to imagine. we looked Which is more important for growth—having a substan- at whether a country’s estimated cognitive skills affected the tial cadre of high performers or bringing everyone up to a basic earnings of immigrants working in the United States. That is. the easier it will be for them to make use of those new impact of improvements at different levels of a nation’s dis. ing parts of the world. the more workers that have at least basic nitive skills and economic growth. In countries on the technological frontier. from as low as dents lacking even basic skills tend to be those that have only 68 E D U C AT I O N N E X T / S P R I N G 2 0 0 8 www. and other innovators To address this question.educationnext. as well as the share of But even if the results seem intuitively correct. can help adapt the new technologies to their countries’ par- tists” at the very top of the distribution who spur economic ticular situation. other but both! When we estimated the importance of each ings if the immigrants were educated in their homeland but within the same model. Some workers need a high level of skill so they tribution of skills. 89 percent of the students achieved at least skills also are ones in which basic skills are near universal. Again. is it a few “rocket scien. To see whether the inclusion of those of high-performing students. achieving students seem to matter. is not one or the home country cognitive skills translated into higher earn. ensuring that the cognitive in Colombia and Morocco to 18 percent in Singapore and skills themselves were not caused by the economic growth. we identified the share of the two measures of cognitive skills are closely related to one students performing at least at a very basic level. are obviously needed. important to economic growth. Most countries that average of all OECD countries.1 percent growth between 1980 and 2000. Even if a country is simply making use of new tech- More Rocket Scientists or Basic Skills for All? nologies developed elsewhere. Once again. we found each of them to be separately not if educated in the United States. Korea and 22 percent in Taiwan. both the perfor- Our commonsense understanding of the importance of mance of countries in ensuring that almost all students achieve good schools can thus be documented quite precisely. have a high percentage of students with very high cognitive try in our study. it seems. The share of students with at least Conversely. is not one or the other but both! We also looked at cognitive skills as measured in the 1960s level. all countries. 6 percent of students performed at that high Is it a few rocket scientists at the very top of the distribution who The answer. most 18 percent in Peru to 97 percent in the Netherlands and of which have both high levels of cognitive skill and rapidly Japan. because importance of basic . countries varied enormously in this through the mid-1980s to see what their impact was on respect. it seems. Finally. thirds of a percentage point every year. petency in mathematics and science. we measured the share of stu. Higher level of performance? The answer. To estimate the should be taken as suggestive rather than definitive. or is it “education for all” that is needed? stantial numbers of scientists. at this very basic level.of the presence in our sample of East Asian countries. no another and our models have difficulty in separating out the more than one standard deviation below the international precise impact of each individually. On average across very powerful. countries with a substantial percentage of stu- basic skills ranged widely among countries. we identified the share of stu- countries in our study influenced our results. albeit diminished. the percentage ranging from as low as 0. But so is a labor force that has the basic dents in each country who reach a threshold of basic com. skills needed to survive in a technologically driven economy. we excluded them dents performing at very high levels—at or above one stan- from one of our models. sub- growth. our basic findings remained intact. Loosely speaking. as is often the case in develop- To gain additional insight into the relationship between cog. A highly at basic levels and their performance in producing high- skilled work force can raise economic growth by about two. engineers. that is. we examined the separate skills. they students who perform at very high levels. In the average OECD coun. technologies. The impact of cognitive skill remained dard deviation over the OECD average.

then. W.5 standard deviations) so The United States has never done well on international assess- that it would score alongside the world leaders? (On average ments of student achievement. had 20-year reform 30 that money effectively raised cognitive skills by 30-year reform Total U. and Fin. a small percentage of highly skilled students. Bush and the nation’s governors together The Economic Benefits of Reform (Figure 4) promised in 1989 the United States would achieve by the year 2000. None of this is meant to suggest that schooling is the only tern is not a perfect one. In other words.educationnext.5 percent increment 40 Percent addition to GDP in GDP is equal to the total the U. www. our growth. at least tentatively. to identify the impact of each type of human capital. Instead. 20 nomic returns to the country would probably have been enough to cover the entire cost of 10 education in 2015 and S P R I N G 2 0 0 8 / E D U C AT I O N N E X T 69 . if a coun. * Figure 4 shows that the benefits of success. These data show that economic benefits from a 1989 reform that raised the U. and Family. that pat. If cognitive skills are so important to economic land as many as 74 points higher. no such gains were realized. the eco. enough to pay for the K–12 education system by 2015. currently performing somewhat below the average of OECD countries. Hong Kong and Taiwan about 60 points higher. After all. there is every reason to believe that the single best route to est” and make sure that “no child is left behind. 0 ful reform grow even more vivid when we look 1989 1994 1999 2004 2014 2019 2024 2029 2034 2039 2044 2049 2054 2059 2064 farther out. K–12 spending the 50 test points that would have brought the United States close to world leadership. Over 75 years. to the highest levels of test performance would cover the cost of K-12 education by 2015. results suggest that GDP would by 2015 be 4.S. so we are able. quality to determine a student’s level of achievement.S.5 percent greater than in the absence of any such Improved GDP from Achieving the Governor's Goals — First in the World by 2000 gains (see Figure 4). and health combine with school we are quite confident that we can recommend that coun. Yet tries both concentrate resources on their “best and bright. the past century.. factor contributing to a society’s cognitive skill development. even a reform that takes effect in 20 years (instead of the governors’ Note: *K-12 education expenditures are assumed to be constant at the level attained in 10 years) yields a real GDP that is 36 percent 2005. or is it “education for all” that is needed? The Impact of Becoming a World Leader education system. its level of cognitive skills on the PISA 2006 math and science exams. most people think that is the sys- What would it mean for economic growth. individual ability. Making American high school students the best in the world in math and Unfortunately.) That increase of 50 test points is exactly what George H. Still. That 4.” higher levels of cognitive skill is strengthening a country’s spur economic growth.S. tem’s primary purpose. science would have had a substantial impact on the nation’s economic But had the promise been fulfilled by 2000. try like the United States. managed to increase An American Exception? its performance by 50 points (or 0. countries such is only about average among the developed countries.S. higher than it would be if there was no change SOURCE: Authors’ calculations in the level of cognitive skills. Yet the as Canada and Korea scored about 50 points higher than the country’s GDP growth rate has been higher than average over U. currently 10-year reform spends on K–12 education.

S. over the 20th century. Yet reformers should bear in mind that that the value of a high-quality education system is substan. In addition. economy.S. if we are to remain economically com- It is also the case that. which will enable them to make better use of ity of its schooling.S.S. education system outpaced the rest of the world. at the University of Munich and heads the Department of education system offer some hope for the future. Part of the answer is that the United States may be resting higher-education system may also be challenged by improve- on its historic record as the world’s leader in educational ments in higher education across the world. Our analyses suggest ics sometimes suggest. We should not.S.S. the governors’ challenge. But. which could eventually have an authors and do not necessarily reflect those of their employers. are doing more to secure property rights and open their some of which are entirely separate and apart from the qual.9 percentage points Economic growth flows only from reforms that bring per year in closed economies.S.S.S. how can we explain the puzzling case of the U. grant universities. maintains generally freer labor their human capital. the expansion petitive. Ludger Woessmann is professor of economics Although the strengths of the U. Taken together. Eliot A. . economy and its higher-education system offer some hope for the future. including lower tax rates and min- imal government production through nationalized indus- tries. Improvements in mathematics performance called for Those economic institutions seem to matter on their own by No Child Left Behind would matter. across the globe. It may be that rich human capital combines with a challenge. Dean T. simply because we The U. Identifying what on international trade. number of years of education their citizens receive. economy Not Just a Matter of Money encourage investment. identified by heavy restrictions actual improvements in cognitive skills. economy and its higher. the situation at the K–12 level should spark concerns about the long-term outlook for the U.? impact on the higher-education system as well. has generally less intrusion of government in the trends could easily accelerate in the coming decades.S.5 percentage points in open works and how to implement it on a society-wide scale remains economies. and direct grants and loans to students. decide that the goals were Higher education expanded with the development of land not legitimate or important. in school attainment has come to an end.S. There of the U. the historic advantage and product markets than most countries in the world. permit the rapid development of new Our evidence of a clear.S. these characteristics of the U. operation of the economy.S. the situation Human Capital and Innovation of the Ifo Institute for Economic at the K–12 level should spark concerns about the long-term Research. higher-education system is Eric A. U. growth. the GI Bill. Babcock & Brown. The U. Put more broadly. colleges and universities rank at the very San Francisco. workers to skills and economic growth should encourage continued reform adjust to new opportunities. money alone will not yield the necessary improvements. strong relationship between cognitive products and activities by firms. The U. in the average less powerful than in many other countries. Jamison is professor of health eco- growth in the U. we need to solve the puzzle of our schools and meet of the U. 70 E D U C AT I O N N E X T / S P R I N G 2 0 0 8 www. Hanushek is a senior fellow at the Hoover Institution of a powerful engine of technological progress and economic Stanford University. have failed to yield actual improvements in student achievement. The extraordinary U.S.S. and allow U.S. We estimate that the expensive attempts around the world to improve schooling effect of a one-standard-deviation improvement in cogni. evaluations. contrary to what crit- and in conjunction with cognitive skills. not only for the U. but also for many nations a laissez-faire economy to foster robust economic growth. and trade unions are of the OECD countries now exceed the U. Most obviously. economies. as half is less government regulation of firms.S. efforts. Many tially diminished in closed economies. By most nomics in the School of Medicine at the University of California. have failed to meet them in the past.educationnext. not accounted for in our analysis. Jamison is an investment professional at top in the world. The opinions expressed in this article are those of the outlook for the U. pushed to open secondary schools to all citizens.Although the strengths of the U. tive skills on annual economic growth is 0. but 2. the United States has other advantages.S. Those the U. Other countries attainment.

such as the Americas. Overall. The essays in this volume study public-private partnerships in different parts of the world. Schools and the Equal Opportunity Problem Ludger Woessmann and Paul E.harvard. using economic and statistical techniques to assess the results of education policy reform in countries including the United States. In recent years. Peterson. School Choice International Rajashri Chakrabarti and Paul E. 2007 Leading researchers from the United States and Europe report on new findings on the effect of education on equal Germany. Peterson.ksg. 2005 In a group of essays originally published in Education Next.harvard. Sweden.htm Program Events: www. Europe. Britain.htm . Rowman & Littlefield Publishers.htm Post Doctoral Opportunities: www. this book examines the likely promise and pitfalls of introducing market concepts to American education. Peterson. MIT Press. there has been a burgeoning of public-private partnerships in different parts of the world.ksg. VISIT US ON THE WEB New Research: eds.harvard. and Italy. ed. Australia and New Zealand.Program on Education Policy & Governance New Books Choice and Competition in American Education Paul E.ksg. Asia. eds. these essays paint the picture of an education landscape that will be greatly shaped by choice and competition in the 21st century. School Choice MIT 2008 INTERNATIONAL Public-private partnerships in education policy play an important role in enhancing the supply as well as the quality of human capital.harvard.